AI Coding change log

Junie Eco mode hands implementation to a lower-cost model

Eco mode uses the selected model to plan and a cheaper model to implement and test; JetBrains reports about half the model cost per task in internal benchmarks. It is generally available in Junie CLI 26.10.20 and later. Date is the post's publication date from the page metadata.

GitHub Enterprise now includes Copilot for every licensed user

Enterprises working with a GitHub or Microsoft account team can get Copilot included with GitHub Enterprise, with AI usage metered against an enterprise budget; users without a GitHub Enterprise license can use Copilot Standalone at $10 per user per month. Self-serve availability is coming in stages.

GitHub Copilot Fusion becomes generally available

Copilot Fusion routes and orchestrates models per task at Standard, High or Max effort, on all Copilot plans, with a limited-time 30% discount on model costs for paid plans.

GitHub Agentic Workflows become generally available

Agentic Workflows, which run coding agents inside GitHub Actions from natural-language Markdown definitions, reach general availability; the default Copilot CLI engine requires a Copilot license.

Amp launches an Android app

Amp's mobile app, previously on iOS and macOS, becomes available on Android, with threads, portals and changes views.

Kiro replaces its upgrade discount with a free first month and 500 trial credits

From November 1, a first upgrade from the free tier gets the rest of that month on Pro free plus 500 trial credits valid for 12 months, and plan fees and credits become prorated; the previous $20 discount ends.

Antigravity 2.28 adds a built-in browser and any-file attachments

The Antigravity App adds a built-in browser in the side panel, a built-in PDF viewer, and attachments of any file type up to 20 MB.

OpenCode 2.2 adds Desktop plugins and a refreshed desktop app

OpenCode 2.2.0 lets users create, install and share experimental Desktop plugins, refreshes the desktop app, and removes the OpenCode GitHub Action.

GPT-6.1 Terra becomes available in Codex

GPT-6.1 Terra, a fast model with image understanding and computer use, becomes available in Codex and ChatGPT Work on Plus, Pro, Business, Enterprise and Edu plans.

Replit keeps published apps running after credits run out, up to an overage limit

Eligible top-up customers keep published apps running at zero credits up to an overage limit; paid Agent pauses, Free mode stays available, and new credits repay the overage first.

Devin Desktop 3.11 rebuilds the Agent Window on the Devin web app interface

Devin Desktop (formerly Windsurf) 3.11.212 brings the Devin web app's sessions sidebar, Automations, Review and Wiki into the Agent Window, with side-by-side sessions and file, change and terminal views.

Amp orbs can run on self-hosted infrastructure

Teams can run an orb provider inside their own infrastructure (Docker, E2B, Kubernetes and others) so Amp spawns, pauses and deletes orbs on compute they control.

Lovable Enterprise customers can group workspaces into an organization

Enterprise companies with several workspaces can manage them in one organization with shared SSO and SCIM, while each workspace keeps its own billing; set up on request through the account team.

Kiro adds memory across sessions in the IDE and CLI

Kiro IDE 1.4.0 and Kiro CLI 2.29.0 add memory that carries preferences, conventions and corrections between sessions; it is off by default.

Cursor's Agent SDK becomes generally available

The Agent SDK lets teams define an agent as a folder of code, test it with evals, and deploy it to Cursor or their own infrastructure.

Amp becomes an agent in Linear

Assigning or mentioning Amp on a Linear issue starts a Linear Agent Session and spawns an orb; the session also appears in Amp's sidebar.

Junie Local supports NVIDIA RTX Spark and its Windows support becomes stable

Junie Local, JetBrains' free on-device coding agent, runs on NVIDIA RTX Spark machines, and Windows support for 24 GB RTX cards moves from experimental to stable.

Replit offers Bring Your Own Cloud for enterprises

Enterprises can run the full Replit platform in their own cloud tenant (private preview), host published apps there, or export images to self-host (early access).

Cline begins rolling out Cline Cloud sessions

Cline 4.1.25 starts a gradual rollout of Cloud sessions: tasks run in a hosted sandbox on a chosen GitHub repo and branch and keep running after the editor closes.

Lovable launches Growth tools for finding customers

Growth tools turn a goal such as a signup target into a plan built from the app, its audience and its traffic, from the same chat used to build.

GPT-5.5 retires from Codex for ChatGPT sign-ins

GPT-5.5 is removed from Codex for users signed in with ChatGPT; API-key sessions and the OpenAI API are unaffected, and users are pointed to GPT-6.1 Sol.

Cursor opens the Origin App Marketplace to public submissions

Origin integrates with over 20 apps for deployments, CI, code review, security, monitoring and project management, and the marketplace opens to public submissions.

Gemini 4 Argon rolls out in Antigravity for Google AI Pro and Ultra

Google brings Gemini 4 Argon to the Antigravity desktop app and CLI for Google AI Pro and Ultra subscribers and eligible Gemini Enterprise subscriptions.

Gemini 4 Argon arrives in Devin Desktop at 75% off through October 27

Cognition adds Gemini 4 Argon to Devin Desktop and the Devin CLI, combining Google's 50% introductory pricing with an extra 50% discount in Devin through October 27, 2026.

Ona Cloud will shut down on October 31, 2026

Following Ona's move to OpenAI, Ona Cloud winds down on October 31, 2026; new sign-ups have closed, October payments and unused credits will be refunded, and users are pointed to Codex Cloud.

OpenHands agents can run on DigitalOcean from Agent Canvas

With OpenHands 1.27.0, developers can create and manage DigitalOcean-hosted agents in isolated microVM sandboxes directly from Agent Canvas.

Bolt moves plan billing to Stripe's Link from November 1

Bolt announces that Link, a Stripe payment service, will bill Bolt plan and domain subscriptions from November 1, 2026; plans and token allowances stay the same, sales tax or VAT is added where required, and checkout may show local currency. Dated to the start of the release-notes week (Oct 12 - 18).

Kiro comes into scope for FedRAMP Class C

Kiro is in scope for FedRAMP Class C in the AWS US East (N. Virginia) Region, opening it to US government workloads with those requirements.

Bolt on mobile adds previews, project management and publishing

Bolt's mobile experience adds full-screen previews, project management, publishing and version restore from a phone; code view, visual edits and the database stay desktop-only. Dated to the start of the release-notes week (Oct 12 - 18).

Google ends sales of new Gemini Code Assist Standard and Enterprise subscriptions

From October 9, 2026, new Standard and Enterprise subscriptions can no longer be purchased; existing subscriptions auto-renew through 2026, auto-renewal ends in 2027, and Google points customers to Antigravity.

Replit Desktop for Windows enters private preview

Replit Desktop can build and run apps locally on Windows PCs in private preview, with each Agent build in a sandbox powered by Microsoft Execution Containers and NVIDIA OpenShell.

Lovable opens invite-only publishing to Free and Pro plans

Free and Pro users can publish an app to an invite-only audience; publishing to a whole workspace or groups still requires Business or Enterprise.

Augment Code sells Cosmos, Auggie CLI and its Context Engine to Harness

Augment Code agrees to sell select assets, including Cosmos, the Auggie CLI and the Code Context Engine, to Harness, and the team behind them moves to Harness.

GPT-6.1 Sol Ultrafast arrives in Codex for Pro $500 and eligible Enterprise and Edu

An Ultrafast mode speeds up GPT-6.1 Sol token generation in Codex, limited to the Pro $500 plan and eligible Enterprise and Edu plans (off by default for Enterprise).

TRAE 3.6 launches the Agent window for multi-agent work

TRAE v3.6.0 adds an agent-centric workspace for orchestrating multiple agents across projects, with cloud tasks and scheduled tasks.

Warp Factories adds Microsoft Teams and Azure DevOps support

Warp Factories can take work from Microsoft Teams channels and from Azure DevOps work items and pull requests.

Claude Code adds Claude Haiku 5.5 as its default Haiku model

Claude Code 2.1.293 adds Claude Haiku 5.5, which becomes the default Haiku model on the Anthropic API, with a 1M-token context window.

v0 turns personal accounts into team workspaces

Personal v0 accounts are migrated to team workspaces, giving every user access to collaboration features; chats, projects and settings move automatically.

Cursor's iOS app can remotely control local agents

Remote Control lets users see and reply to agents running on their own computer from the Cursor iOS app; it is on by default except for Enterprise organizations.

Cline for Desktop cloud sessions get the same models as Cline on the web

Cloud sessions can use Cline Usage-Billing, ClinePass, and ClineFree models, and a new cloud session waits until a model is ready before sending.

Lovable apps can add live voice conversations with GPT Live

Apps built with Lovable can add live voice conversations to their AI features with GPT Live; Lovable sets up the conversation you describe.

Codex CLI 0.160 adds opt-in Guardian review context and task history browsing

Browse older tasks in the agent command center, start sessions outside a project with workspace defaults, and opt in to Guardian review capabilities that include context from agent handoffs.

Junie adds Junie Lite, a free AI layer

Junie CLI release 26.9.22 (3419.26) adds Junie Lite — a free AI layer with no trial and no card.

Lovable Enterprise workspaces can keep AI processing in the EU

Workspace admins and owners on Enterprise plans can enable EU inference, which runs the workspace's AI model requests on providers in the European Union.

OpenCode 1.18.34 signs macOS CLI binaries with a Developer ID

Sends namespaced session identity headers with model requests and signs macOS CLI release binaries so they run reliably on newer macOS versions.

Antigravity 2.19 lets you message subagents directly

Send messages straight to a subagent from the message box, export rendered Markdown as a PDF, and revert only the conversation when you undo.

Zed 1.22 adds model choice for subagents and BYOK for GPT-6.1 Sol

This week's release adds BYOK support for GPT-6.1 Sol, on-the-fly model selection for new subagents, and automatic subagent compaction.

Kiro IDE 1.2 runs multi-agent Workflows in the background

IDE 1.2 introduces Workflows, strengthens safeguards for untrusted workspaces, and adds enterprise controls for sign-in methods.

Bolt.new acquires Dokai, its first acquisition

Bolt.new announces the acquisition of Dokai, a startup building autonomous AI agents for enterprise sales workflows; the Dokai team joins Bolt's AI and machine learning organization.

GPT-6.1 Sol becomes available in GitHub Copilot

OpenAI's GPT-6.1 Sol is added to the models available in Copilot.

Sign in with ChatGPT bills GPT usage to a ChatGPT Plus or Pro plan (v3.10.48)

Eligible ChatGPT Plus or Pro subscribers can bill GPT model requests in Devin Desktop to their ChatGPT plan.

Codex CLI 0.159 lets new input steer Codex mid-response

Opt-in instant_interrupt lets new input steer Codex during model responses or long-running code-mode calls; new sessions get a compact welcome screen.

Antigravity 2.18 adds a plugin marketplace

A new Customizations tab and marketplace for discovering, installing, and managing plugins.

Claude Sonnet 5.5 added as the default Sonnet model (v2.1.284)

Release v2.1.284 adds Claude Sonnet 5.5, now the default Sonnet model on the Anthropic API.

Claude Sonnet 5.5 becomes available in GitHub Copilot

Anthropic's Claude Sonnet 5.5 is added to the models available in Copilot.

Kiro CLI 2.25 adds Power management and V3 output styles

Install and uninstall Powers from chat or the command line, choose V3 output styles, and automate session teardown with a new Hook.

Build apps for Meta devices, plus newer OpenAI and Anthropic models

The weekly update adds building for Meta devices, app creation through Muse, inline data charts, and new OpenAI and Anthropic models.

Amp runners can be shared with a whole workspace

Start a runner with --share and everyone in the workspace can start threads on that machine from ampcode.com.

New default policy for Copilot features arrives for Business and Enterprise

Admins can set a global default policy for generally available Copilot features; for 28 days it is configurable without affecting user access.

Zed 1.21 adds BYOK for Claude Opus 5.5 and GPT-6 models

Adds BYOK support for Claude Opus 5.5 and GPT-6 Astra, Sol, and Luna, a language server command picker, and a setting that keeps the system awake during long agent turns.

Gemini CLI v0.61 hardens against indirect prompt injection

Core security hardening prevents indirect prompt injection via build file modifications and untrusted flags, and hardens filesystem boundaries.

Claude Code cloud sessions become generally available

Cloud sessions, previously Claude Code on the web, are now generally available for Pro, Max, Team, and eligible Enterprise users.

Rollouts and Security Review bots launch on Teams and Enterprise

Rollouts monitors each pull request as it deploys and reports per-environment health; Security Review posts exploitable-bug findings on every PR.

OpenHands 1.23 ships a universal macOS desktop build

The desktop app ships as a universal macOS DMG with per-architecture bundled runtimes.

Junie CLI adds Claude Opus 5.5, GPT-6 Sol and Luna, and Grok 4.7

Junie CLI release 26.9.22 (3419.7) adds support for Grok 4.7, Claude Opus 5.5, and GPT-6 Sol and GPT-6 Luna.

v0 widens model choice, with Enterprise owner controls

An updated model picker lets users browse and add AI Gateway models; Enterprise team owners control whether custom models are allowed.

Claude Opus 5.5 added as the default Opus model (v2.1.280)

Release v2.1.280 adds Claude Opus 5.5, now the default Opus model.

Lovable adds a free daily chat allowance on Free, Pro and Business plans

Lovable launches workspace and in-project chat, free within a daily allowance on Free, Pro and Business plans; beyond it, each message is priced on the work done, typically a fraction of a credit.

AGENTS.md support: read when a project has no CLAUDE.md (v2.1.277)

Release v2.1.277: in a project with no CLAUDE.md, Claude Code reads AGENTS.md instead.

Custom API connectors for Agent enter beta on Pro and Enterprise

Teams can connect Agent to APIs outside Replit's integration library by adding an endpoint and authentication.

Factory adds Factory Private for VPC, on-premises and air-gapped deployment

Factory introduces two deployment models, Factory Managed and Factory Private, the latter for VPCs, on-premises and air-gapped networks.

Warp adds native shell completions and fzf and atuin hand-off

Trigger fzf (ctrl-r or ctrl-t) and atuin from Warp, and get native shell completions alongside Warp's built-in completion specs.

Bolt introduces Bolt Lite, a $9/month Forge-only plan

Alongside the Bolt Forge research preview, Bolt introduces Bolt Lite, a standalone $9 per month plan with Forge-only access, offered via a waitlist.

Bolt Forge, an agent on open-source models, starts a research preview

Bolt Forge joins Standard and Max as a third agent running open-source models with a larger usage allowance; it is included with individual Pro plans, and a $9/month Forge-only Bolt Lite plan opens a waitlist.

Augment's IntelliJ plugin asks before reading sensitive files

IntelliJ plugin v0.491.0: the agent asks for approval before reading .env files, private keys, certificates, and credential-named files, and MCP over HTTP now uses the configured proxy.

Cline releases Cline Desktop, an open-source app for open-weight models

Cline releases an early version of Cline Desktop, an open-source desktop app focused on open-weight models, available for Mac with a Windows beta.

Amp adds a free Hobby tier for bring-your-own compute and models

Amp becomes free to use with your own compute and model subscriptions or keys, introducing a free Hobby tier and removing BYOK token fees outside Enterprise; it charges for remote orbs.

claude plugin eval runs a plugin's eval suite with scored results (v2.1.269)

Release v2.1.269 adds `claude plugin eval` for reproducible, scored plugin evaluations.

Fusion harness comes to Devin Desktop and CLI

Cognition makes its Fusion model harness available in Devin Desktop and the Devin CLI.

Replit becomes usable from desktop MCP clients

Through the Replit MCP server, compatible desktop clients can create, inspect, update, and publish Replit apps.

v0 team chats become visible to the team by default

New chats created in a team workspace are visible to the team by default; owners can change the default visibility.

Projects launches in beta for long-running, multi-agent work

A coordinator agent plans larger bodies of work and delegates to cloud and local agents, keeping shared context across months. Rolling out to all users.

Remote agent hosts: connect to agents over SSH (v3.10.23)

SSH hosts can be saved and selected from the location picker, and sessions recover their transcript after reconnecting.

Factory Pro plan adds managed computers with five compute hours a month

Factory's Pro plan now includes managed computers with five compute hours each month.

Warp adds first-class support for Grok Build sessions

Warp detects Grok Build sessions, adds branded chrome in the footer, and enables rich input mode.

Cascade is removed; Devin Local becomes the only agent (v3.9.19)

Existing Cascade conversations can be migrated with the Continue in Devin Local prompt.

Scheduled daily backups for production databases

Scheduled backups create one full restore point per day, with a configurable retention period.

Gemini Code Assist stops new console purchases for billing accounts without a subscription

New Gemini Code Assist subscriptions can no longer be bought through the Google Cloud console by billing accounts that do not already have an active subscription.

Copilot Business and Enterprise sign-ups reopen with upfront seat payment

GitHub began gradually reopening Business and Enterprise sign-ups for card and PayPal customers, now requiring payment per seat before access.

Self-hosted machines keep agent tool execution inside your network

Codebases, build outputs, and secrets stay on internal machines; team pools scale self-hosted workers with demand.

Claude Fable 5.1 added as the default Fable model (v2.1.257)

Release v2.1.257 adds Claude Fable 5.1, now the default Fable model.

Codex in Ona follows the organization's configured model access

Inherited Codex access tokens no longer override the model access configured for the organization.

Intelligent model routing picks model and effort level automatically

Selecting Auto lets Replit route each part of the work to the model and effort level that balance quality and cost.

Cloud agents can start from scratch without a connected repository

Work is saved to a Cursor Origin repo; live previews forward to the browser, with optional publishing through Vercel.

Codex in Ona gives clearer error messages

When a run cannot continue, messages focus on what happened and what to do next, with a reference ID tied to the support bundle.

JetBrains launches free on-device Junie Local for Mac

Junie Local runs a tuned local model entirely on the user's Mac with no credits or quota and no code leaving the machine.

Gemini Code Assist discontinues first-month subscription credits

Google discontinues the credits for the first month of Gemini Code Assist usage; customers already receiving them are unaffected.

Antigravity included in Gemini Enterprise Standard and Plus licenses

Google bundled Antigravity into eligible Gemini Enterprise subscriptions and added IDE extensions.

Cloud agents gain event subscriptions, custom modes, and per-subagent VMs

Agents can watch PRs, Slack threads, or schedules and wake on events; subagents can run on their own virtual machines.

SpaceX completes acquisition of Cursor

Cursor announced it has officially been acquired by SpaceX, completing a process that began with an April partnership with SpaceXAI.

OpenAI completes its acquisition of Ona

An update to Ona's announcement states the transaction closed on August 10, 2026 and Ona is now part of OpenAI.

Ona Cloud drops the Anthropic harness on Ona-managed model access

Ona says the Anthropic harness is no longer available with Ona-managed model access, effective immediately, and migrates existing automations to equivalent OpenAI models; enterprise BYOK customers can keep their own Anthropic keys for now.

Bolt retires its v1 Agent and Discussion Mode

Bolt retires the legacy v1 Agent and Discussion Mode, automatically switching remaining v1 projects to Bolt Agent, after blocking v1 for new projects from April 13, 2026.

Replit cuts Cloud deployment and storage prices on all plans

Replit Cloud prices dropped across plans from billing cycles starting on or after August 1, 2026; autoscale compute fell from $3.20 to $0.60 per million units.

Cursor launches Cursor Start plan for India at ₹649 per month

Cursor introduced Cursor Start, an INR-priced plan for developers in India payable with UPI, including access to Grok 4.5 and Composer.

Amp launches monthly subscriptions in beta

Amp introduces monthly subscriptions in beta alongside its existing pay-as-you-go token pricing, with included usage of its remote 'orbs' compute.

Cline launches ClinePass, a $9.99/month subscription for open-weight models

Cline introduces ClinePass, a monthly subscription bundling curated open-weight models with higher API rate limits across its IDE, CLI and SDK surfaces, priced at $9.99 per month.

Gemini Code Assist IDE extensions stop serving free and Google AI Pro/Ultra users

The IDE extensions stop serving requests for Google AI Pro and Ultra and free Gemini Code Assist for individuals; organizations on Standard or Enterprise licenses keep access.

Lovable merges build, Cloud and AI usage into one credit balance

Lovable replaces separate dollar balances for Cloud and AI with a single credit balance, converting remaining balances at each plan's credit rate; it states Cloud and AI costs are unchanged.

Gemini CLI stops serving free and Google AI Pro/Ultra users

Gemini CLI and the Gemini Code Assist IDE extensions stop serving requests for Google AI Pro and Ultra and free-of-charge users; organizations on Code Assist Standard or Enterprise keep access.

GitHub Copilot desktop app becomes generally available

The GitHub Copilot app for agent-driven development is now generally available on macOS, Windows and Linux.

Junie coding agent leaves beta

JetBrains declared Junie generally available across the IDE and terminal.

Sign-ups for Copilot individual plans reopen

New sign-ups for individual Copilot plans, paused at the June 1 billing transition, reopen.

Factory announces Factory 2.0, moving from coding agents to software factories

Factory announces Factory 2.0, describing a shift from individual coding agents to organization-wide, agent-native systems covering the development lifecycle.

Kiro adds Pro Max plan at $100 per month

Kiro introduced Pro Max, a $100/month tier with 5,000 credits to fill the gap between Pro+ ($40) and Power ($200).

Ona agrees to join OpenAI as part of the Codex team

Ona announces an agreement to join OpenAI's Codex team, with both companies remaining independent until close.

TRAE SOLO is renamed TRAE Work

Trae renamed SOLO to TRAE Work, positioning it as an AI workspace for all professional roles across desktop, web and mobile.

Windsurf is renamed Devin Desktop

Cognition introduces Devin Desktop as the next generation of Windsurf, with the Agent Command Center as the default surface and the IDE kept backwards-compatible.

Cursor adds Premium Teams seat and first-party usage pool

Teams seats gain a separate first-party models pool, and a new Premium seat offers 5x Standard usage at $96/seat/month annual or $120 monthly.

Usage-based billing goes live for all Copilot plans; Max launches

Usage-based billing is active for every Copilot plan, Copilot code review also consumes Actions minutes, and Copilot Max launches for power users.

Warp opens BYOK to the Free plan and makes it free for small teams

Warp extends bring-your-own-key and custom inference endpoints to the Free plan, free for individuals and companies of 10 or fewer people. Larger companies must use Business or Enterprise, where BYOK usage consumes platform credits at a lower rate.

Google announces Gemini CLI transition to Antigravity CLI

Google said Gemini CLI and Code Assist IDE extensions would stop serving Google AI Pro, Ultra and free individual users on June 18, 2026; enterprise licenses unaffected.

Google adds $100 per month AI Ultra tier for Antigravity users

A new $100/month Google AI Ultra plan sits between Pro and top-tier Ultra, with 5x the Pro plan's token rate limit.

Google releases Antigravity 2.0 standalone agent desktop app

Antigravity 2.0 is a separate agent-first desktop application for macOS, Linux and Windows with no IDE, adding dynamic subagents and async tasks.

Google cuts top Google AI Ultra plan for Antigravity from $250 to $200

Google reduced the top-tier AI Ultra price to $200/month, merged Gemini Flash and Pro limits into one pool, and made AI credits an overage-only mechanism.

New Max plan and flex allotments announced for individual plans

GitHub announces a new Max plan and flex allotments for Pro and Pro+, effective with the June 1 lineup.

Cursor moves Bugbot from $40 seat fee to usage-based billing

Bugbot drops its $40 per seat monthly subscription for usage-based billing on Teams and Individual plans, effective at renewals after June 8, 2026.

Zed launches Zed for Business at $30 per seat

Zed introduced a business plan with org-wide privacy controls and admin controls, generally available at $30 per seat per month.

Anthropic doubles Claude Code five-hour rate limits on paid plans

Citing new compute from a SpaceX agreement, Anthropic doubled Claude Code's five-hour limits for Pro, Max, Team and seat-based Enterprise and removed the peak-hours reduction.

Factory renames its Pro Plus plan to Plus

Factory's changelog records that the Pro Plus individual plan is renamed Plus throughout the product.

AWS announces Amazon Q Developer end of support and halts new sign-ups from May 15

AWS announces that Amazon Q Developer IDE plugins and paid subscriptions reach end of support on April 30, 2027, and that new sign-ups stop on May 15, 2026. Customers are directed to migrate to Kiro.

Zed reaches version 1.0

Zed declared version 1.0 of its Rust-built editor, citing broad language, Git, SSH remoting and debugger coverage across platforms.

Copilot announces a move to usage-based billing from June 1

All Copilot plans transition to usage-based billing, with usage consuming GitHub AI Credits.

GitHub pauses self-serve Copilot Business sign-ups

New self-serve Copilot Business sign-ups were paused for organizations on GitHub Free and Team plans; existing customers were unaffected.

GitHub pauses new sign-ups for Copilot Pro, Pro+ and Student

New sign-ups for the Student, Pro and Pro+ plans were paused to prioritize service for existing customers; Copilot Free stayed open.

Copilot tightens individual plan limits and removes Opus from Pro

GitHub tightened usage limits on individual plans, with Pro+ offering more than 5x Pro's limits, and removed Opus models from the Pro plan.

Windsurf 2.0 bundles Devin cloud agent into self-serve plans

Windsurf 2.0 made the Devin cloud agent available inside the editor, included with every self-serve plan and billed against existing quota.

GitHub pauses new Copilot Pro free trials

Citing abuse of the trial system, GitHub paused new Copilot Pro trials; an April 13 update extended the pause to existing trials.

Windsurf removes daily limits for Max and adds Adaptive router

Responding to feedback that new plans were too restrictive, Windsurf removed the daily limit for Max users and launched an Adaptive model router billed at a fixed per-token rate.

Amp removes ads from Amp Free and pauses free grants for some users

Amp stops showing ads in Amp Free; most users keep the daily free grant without ads, while it is paused for some less-active users.

Windsurf launches $200 per month Max plan

Alongside the quota change, Windsurf added a Max plan for power users at $200/month; Pro lists at $20/month and Teams at $40/seat.

Windsurf replaces credits with daily and weekly usage quotas

Windsurf announced that Free, Pro and Teams plans move from credits to daily and weekly quotas, effective March 19, 2026, with extra usage billed at API pricing.

Replit launches Agent 4

Replit introduced Agent 4, adding design variants on an infinite canvas and background task handling.

Gemini 3.1 Pro becomes available in Jules

Jules adds Gemini 3.1 Pro as an available model.

Replit cuts Core plan price from $25 to $20 per month

The Core plan's price dropped to $20/month from $25/month and now allows inviting up to 5 people to a workspace.

Replit launches $100 Pro plan and sunsets Teams

Replit introduced a Pro plan at $100/month with tiered credits and up to 15 builders, and sunset Teams, upgrading Teams users to Pro.

Trae moves to token-based pricing from February 24, 2026

Trae announced a switch to token-based pricing effective February 24, 2026, with on-demand usage billed at API rates beyond plan allowances.

Trae adds Lite, Pro+ and Ultra tiers from $3 to $100 per month

Alongside token pricing, Trae added Lite ($3), Pro+ and Ultra ($100) tiers to its existing Pro plan, plus a 14-day Pro trial for new users.

Enterprise plans with Claude Code become self-serve

Organizations can now buy Claude Enterprise directly online without a sales conversation; the single seat type includes Claude, Claude Code, and Cowork.

Cursor splits individual plans into first-party and API usage pools

Individual plans now have a first-party models pool with higher limits for Auto and Composer 1.5, and a separate API pool billed at model API prices.

OpenAI launches the Codex desktop app for macOS

The Codex app is a macOS desktop interface for running parallel agent threads with worktrees, Git tooling, skills and automations.

Codex opens to ChatGPT Free and Go and doubles paid rate limits

For a limited time, Free and Go plans include Codex and paid plans get double rate limits across app, CLI, IDE and cloud.

Claude Code access added to every Team plan Standard seat

Claude Code, previously tied to premium seats on Team plans, is now included with every Team plan standard seat.

Amp offers up to $10/day in ad-supported free credits for all modes

Every Amp user can receive ad-supported free credits replenishing hourly, up to $10 per day, usable with the full smart agent.

Gemini CLI free tier gains access to Gemini 3 Pro and Flash

All free-tier Gemini CLI users gained access to Gemini 3 Pro and Flash via the Preview Features setting.

OpenHands 1.0.0 ships on the new software-agent-sdk

OpenHands releases version 1.0.0, rebuilt on its new software-agent-sdk, and moves the OpenHands CLI to its own repository.

Amp spins out of Sourcegraph as Amp Frontier Corporation

Amp becomes a separate company, spinning out of Sourcegraph as an independent research lab.

Google launches Antigravity in free public preview

Google introduced Antigravity, an agent-first development platform, available in public preview at no charge with Gemini 3 Pro rate limits.

Kiro becomes generally available

Kiro exited preview and became generally available, with team features across the IDE and terminal.

Amazon Q Developer CLI is succeeded by Kiro CLI

AWS launches Kiro CLI, which carries over the Q Developer CLI's agent functionality; AWS documentation states the Amazon Q Developer CLI has been rebranded to Kiro.

v0 gives paid users $2 in free credits every day

v0 adds a daily free credit grant of $2 for all paid users.

TRAE SOLO coding agent becomes generally available

Trae's SOLO responsive coding agent left beta, adding SOLO Coder with plan mode and sub-agent orchestration.

OpenHands Cloud adds a free bring-your-own-LLM tier

OpenHands opens bring-your-own-key to all Cloud users through a free BYOK tier limited to one user, with daily and parallel conversation limits.

v0 launches a $100/user/month Business plan

v0 adds a Business plan between Team and Enterprise at $100 per user per month, with training opt-out by default, shared credit purchasing and centralized billing.

OpenCode 1.0 ships with rewritten TUI

OpenCode 1.0 replaced its Go/Bubbletea terminal UI with an in-house OpenTUI framework written in Zig and SolidJS.

Warp replaces Pro, Turbo and Lightspeed with a $20 Build plan and credit reloads

Warp introduces a single usage-based Build plan at $20 per month with 1,500 credits, plus reload credits and bring-your-own-key, and deprecates Pro, Turbo and Lightspeed. Existing customers move over at their first renewal after December 1, 2025.

Codex adds on-demand credits for ChatGPT Plus and Pro

Plus and Pro users can buy on-demand credits for Codex usage beyond their plan's included limits.

Cursor 2.0 launches with Composer model and multi-agent interface

Cursor 2.0 introduced Composer, its first in-house coding model, and an agent-centric interface for running multiple agents in parallel.

Augment Code switches all plans to credit-based pricing

Augment Code's credit-based pricing goes live, replacing user-message plans; new sign-ups move first and existing paid users migrate between October 21 and 31 at the same dollar value.

Cline launches Enterprise and sets Teams pricing at $20 per user after 10 seats

Cline announces Cline Enterprise with sales-quoted pricing and a Teams tier that is free through 2025, then free for up to 10 users and $20 per user per month beyond that.

Claude Code on the web launches as research preview

Anthropic introduced Claude Code on the web, letting users delegate coding tasks from the browser to sessions on Anthropic-managed cloud infrastructure.

Zed becomes available on Windows

Zed shipped a stable Windows release, making Windows a fully supported platform with weekly updates.

Amp launches Amp Free, an ad-supported free mode

Amp adds a free mode funded by advertising from developer and infrastructure companies, using a mix of open-source and frontier models.

Codex becomes generally available

Codex reached general availability, adding @Codex in Slack, the Codex SDK and new admin tools.

Lovable launches Lovable Cloud and Lovable AI

Lovable launches Lovable Cloud for built-in databases, authentication and storage, and Lovable AI for adding AI features to apps by prompting, with free monthly AI credits on every plan.

Jules launches Jules Tools, a command-line interface

Google launches Jules Tools, a CLI for controlling the Jules agent from the terminal; the Jules API follows on October 3, 2025.

Copilot coding agent becomes generally available

GitHub's asynchronous Copilot coding agent, which works via GitHub Actions and opens draft pull requests, is now generally available to all paid Copilot subscribers.

Zed moves hosted AI from prompt limits to token-based pricing

Zed switched Zed AI from prompt-based limits to token-based pricing, live immediately for new users with existing users migrated over three months.

Google AI Pro and Ultra plans gain higher Gemini Code Assist limits

Google AI Pro and Ultra subscribers receive Gemini CLI and Gemini Code Assist with higher model request limits.

OpenHands Cloud adds a $20/month subscription and drops model markup

OpenHands Cloud introduces a $20 per month subscription that unlocks model choice at API cost without markup and bring-your-own-key; previously it charged 2x API costs for Claude Sonnet.

Cline becomes available natively in JetBrains IDEs

Cline announces a native extension for JetBrains IDEs including IntelliJ IDEA, PyCharm, WebStorm, Android Studio, GoLand and PhpStorm.

GitHub to remove $0 premium request budgets for enterprise and team accounts

GitHub announced removal of account-level $0 Copilot premium request budgets for older enterprise and team accounts, later scheduled to begin December 2, 2025.

Kiro replaces vibe and spec limits with a unified credit pool

Kiro moved to unified credits (Free 50, Pro $20/1,000, Pro+ $40/2,000, Power $200/10,000) with $0.04/credit overages, billing from October 1.

Augment Code introduces a $20/month Indie plan

Augment Code adds an Indie plan at $20 per user per month with 125 user messages.

Replit launches Agent 3 with in-browser app testing

Replit introduced Agent 3, which tests apps in a browser, fixes issues automatically, and can generate other agents and automations.

Gitpod Classic pay-as-you-go is set to sunset on October 15, 2025

Ona announces that Gitpod Classic pay-as-you-go users cannot log in or start environments after October 15, 2025 and must migrate to Ona; Enterprise customers are excluded.

Gitpod re-founds as Ona, an agent platform

Gitpod announces it is re-founding the company as Ona, positioned as mission control for a personal team of software engineering agents.

Trae adds token-metered Max Mode billed against Fast Requests

Trae introduced Max Mode for Pro users, metering tokens per session and converting cost into Fast Request equivalents.

JetBrains ties AI quotas, including Junie, to dollar-valued AI credits

From August 25, 2025, JetBrains AI plans grant AI Credits equal to plan price, reducing effective quotas, with paid top-ups beyond the included amount.

Team and Enterprise plans add premium seats with Claude Code

Anthropic introduced a premium seat tier for Team and Enterprise plans offering more usage and access to Claude Code.

Kiro pricing plans go live; preview users move to Free tier

Kiro activated paid plans to clear its waitlist; eligible accounts moved to a Free tier of 50 vibe and 0 spec requests per month.

Cursor moves Teams to variable request costs and meters Auto usage

From renewals after September 15, 2025, Teams agent requests are charged by work done rather than fixed credits, and Auto counts toward individual plans' included usage.

v0.dev is renamed v0.app

Vercel renames v0.dev to v0.app and positions it as an agentic AI builder for everyone.

Lovable introduces a Business plan and yearly billing

Lovable adds a Business plan for teams with SSO, restricted projects and templates on top of Pro, and makes yearly subscriptions available for paid plans.

Jules exits beta and launches publicly

Google takes Jules out of public beta and launches it publicly, powered by Gemini 2.5.

Jules adds Google AI Pro and Ultra tiers with 5x and 20x limits

With the public launch, Jules introduces structured tiers: Google AI Pro subscribers get 5x higher limits and Ultra subscribers 20x higher limits.

AWS introduces Kiro agentic IDE in preview

Kiro launched as a spec-driven agentic IDE in preview.

Cognition signs a definitive agreement to acquire Windsurf

The acquisition covers Windsurf's IP, product, trademark and brand, and team.

Cursor apologizes for Pro pricing rollout and offers refunds

Cursor said its June changes to individual plans were not communicated clearly and committed to refunding unexpected usage charges from the prior three weeks.

Lovable folds its Teams plan into Pro and makes collaboration free

Lovable makes workspace collaboration free on Free and Pro plans for up to 20 members and moves Teams subscribers to the Pro plan, which it says costs 20% less from the next renewal.

Google launches open-source Gemini CLI with free tier

Google released Gemini CLI as an open-source terminal agent, free with a personal account at 60 requests/minute and 1,000/day during preview.

Warp 2.0 launches as an Agentic Development Environment

Warp releases Warp 2.0, repositioning the product as an Agentic Development Environment built around agent workflows. The same release raises AI usage limits on Pro and Turbo and adds pay-as-you-go overages.

Copilot begins enforcing monthly premium request allowances

Premium request allowances for Copilot Pro, Pro+, Business and Enterprise are now enforced, with optional pay-per-request spending beyond the allowance.

Replit moves Agent to effort-based pricing

Replit replaced its flat $0.25-per-checkpoint Agent pricing with effort-based pricing tied to the time and compute each request uses.

Cursor launches Ultra plan at $200 per month

Cursor introduced an Ultra tier priced at $200/month offering 20x the usage of Pro, aimed at power users wanting predictable costs.

Cursor Pro moves from request limits to compute-based usage

The Pro plan switched by default from a 500-request limit to compute limits worth at least $20 of model inference at API prices per month, with unlimited Auto.

Cursor 1.0 ships with Bugbot and Background Agent for all users

Cursor 1.0 added Bugbot automated PR review, memories in beta, one-click MCP setup, Jupyter support, and general availability of Background Agent.

Codex becomes available to ChatGPT Plus users

OpenAI extended Codex to ChatGPT Plus subscribers and added optional internet access during tasks.

Jules raises beta limits to 60 tasks per day and 5 concurrent

Google raises Jules usage limits during the beta to 60 tasks per day with 5 concurrent tasks.

Factory makes Droids generally available

Factory announces general availability of its Droids across the software development lifecycle, with team pricing of $40 per team plus $10 per active user per month.

Claude Code becomes generally available alongside Claude 4

Anthropic made Claude Code generally available with the Claude 4 launch, adding beta VS Code and JetBrains extensions and a Claude Code SDK.

OpenAI launches Codex cloud agent research preview

OpenAI introduced Codex, a cloud software engineering agent powered by codex-1, initially for ChatGPT Pro, Business and Enterprise users.

v0 switches from message counts to token-metered credits

Vercel moves v0 to usage metered on input and output tokens converted to credits, with $5, $20 and $30 per user of monthly included credits on Free, Premium and Team.

Zed launches Pro plan with 500 prompts for $20 per month

With its agentic editing launch, Zed introduced hosted AI plans: 50 prompts/month free and a new Pro plan with 500 prompts for $20/month.

Augment Code moves to per-user-message pricing

Augment Code replaces request and tool-call based charging with pricing per user message sent to its agent; the Developer plan is set at $50 for 600 user messages.

Amazon Q Developer brings its agentic coding experience to VS Code

AWS makes an interactive agentic coding experience available in the Amazon Q Developer extension for Visual Studio Code, at no additional cost for both Free and Pro tier users.

Windsurf raises Free tier limits and enables Cascade write mode

The Free tier gained higher limits, including Cascade in write mode, 25 prompt credits per month, and unlimited Fast Tab.

Windsurf drops flow action credits in favor of prompt credits

Windsurf removed Flow Action Credits effective April 21, 2025; plans now use prompt credits consumed per message rather than per tool call.

Junie becomes available in JetBrains IDEs under one AI subscription

JetBrains made Junie and AI Assistant available in its IDEs under a single subscription that includes a free tier.

GitHub launches Copilot Pro+ individual tier

GitHub introduced Copilot Pro+, an individual tier with access to the latest models, priority previews, and 1,500 premium requests per month.

Warp becomes available on Windows

Warp's changelog announces that the terminal is now available for Windows, extending support beyond macOS and Linux.

Google launches free Gemini Code Assist for individuals

Google makes Gemini Code Assist for individuals globally available at no cost, with up to 180,000 code completions per month.

Gemini Code Assist adds up to 50 free license credits for new customers' first month

Google applies up to 50 license credits for the first month to new billing accounts that have never had a Gemini Code Assist subscription.